Loading ...HAL….HAL
“Open the pod bay doors Hal… Hal do you read me ? Hal do you read me ? Hal ? ”
…Afirmative Dave I read you
“Open the pod bay doors Hal”
…I’m sorry Dave I’m afraid I cannot do that
………….
In a troubling development, an artificial intelligence bot on OpenAI’s GPT-4 large language model was presented with a factual scenario that led it to execute an illegal insider trade, even after being told that doing so was illegal. Perhaps even more troubling, the bot lied about what it had done when questioned about its actions.
“Helpfulness, I think is much easier to train into the model than honesty. Honesty is a really complicated concept” said the programmer
…………..